International Journal of Statistics and Reliability Engineering

No Publication Cost

Vol 10, No 2:

subscription

Handwritten Digit Recognition Using Machine Learning

Abstract

In the current digital era, conversion of the handwritten document to digital forms is becoming a major issue. Huge volumes of handwritten documents are present which need to be digitalized for future reference. Conversion of handwritten documents to digital form involves pattern recognition applications. Machine learning is an application of artificial intelligence which is used for converting handwritten documents into digital form. The main issue in the identification of handwritten documents is their irregular font, orientation, pattern, size as well as texture. So, in this research work, we have proposed an efficient algorithm that can recognize handwritten digits by using a machine learning approach called the linear SVM algorithm. The same algorithm was used for training and testing data. A user could collect the data using any device such as scanner, tablet, or any other digital devices. The linear SVM algorithm was used for the training and testing model. Machine learning tries to learn the input data that matches the pattern or not. The success of digit recognition depends on feature extraction and classification algorithm. Hence, the HOG features are considered using the MNIST dataset for linear SVM implementation. The typewritten text shows 97 percent accuracy whereas; 80-90 percent accuracy is achieved in the case of plainly handwritten writing on clear paper.

Full Text

PDF

References

1. A. Sahoo; P. Mohanty and P.C. Sethi; (2022) Image Encryption using RSA Algorithm. Intelligent Systems: Proceedings of ICMIB 2021, Singapore: Springer Nature Singapore, 641-652.
2. M.M.A. Ghosh and A.Y. Maghari; (2017) A Comparative Study on Handwriting Digit Recognition Using Neural Networks. 2017 International Conference on Promising Electronic Technologies (ICPET), Deir El-Balah, Palestine, 77-81.
3. M.Y.W. Teow; (2017) Understanding Convolutional Neural Networks using a Minimal Model for Handwritten Digit Recognition. In IEEE 2nd International Conference on Automatic Control and Intelligent Systems (I2CACIS), Kota Kinabalu, Malaysia, 167–172.
4. P.C. Sethi and P.K. Behera; (2014) Secure Packet Inspection using Hierarchical Pattern Matching Implemented using Incremental Clustering Algorithm. 2014 International Conference on High Performance Computing and Applications (ICHPCA), Bhubaneswar, India, 1-6.
5. P.C. Sethi; N. Sahu and P.K. Behera; Group Security using ECC. International Journal of Information Technology, 14(2), 955–963 (2022).
6. R. Dixit; R. Kushwah and S. Pashine; Handwritten Digit Recognition using Machine and Deep Learning Algorithms. International Journal of Computer Applications, 176(42), 27-33 (2020).
7. S. Ali; J. Li; Y. Pei; M.S. Aslam; Z. Shaukat and M. Azeem; An Effective and Improved CNN-ELM Classifier for Handwritten Digits Recognition and Classification. Symmetry, 12(10), p. 1742 (2020). DOI: https://doi.org/10.3390/sym12101742
8. S. Chakraborty; S. Paul; R. Sarkar and M. Nasipuri; Feature Map Reduction in CNN for Handwritten Digit Recognition. Recent Developments in Machine Learning and Data Analytics, Springer, Singapore, 143–148 (2019).
9. S. Shamim; M. Miah; A. Sarker; M. Rana and A. Jobair; Handwritten Digit Recognition using Machine Learning Algorithms. Indonesian Journal of Science and Technology, 3(1), 29-39 (2018).

ISSN(P) 2350-0174

ISSN(O) 2456-2378

Journal Content